get bad score running the sample code

[Aun0124]

1. I configure everything and run the whole script and get negative score on the boston datasets. Is it just a sample since i get a bad score is normal ?

2. The default is only using gradient boosting for the classification and regression and not automatically choose the best model for taining and prediction?